Structure Learning Training and Cognitive Flexibility

NCT05611788 | Completed

• 1 other identifier: IRB-2021-03-058
• Study Type: interventional
• Target: 108
• Locations: 1 country
• Sites: 1

Brief Summary

The overarching aim of the project is to assess whether structure learning is an effective tool to train and improve cognitive flexibility and whether this is transferable to learning and other cognitive skills. The investigators will employ a multi-modal approach that combine both cognitive-behavioural and neuroimaging measures to examine how functional brain activations and/or cognitive performances are affected when participants go through a structure learning training paradigm relative to a working memory training paradigm (active control) or a no-training paradigm (passive control).

Conditions

Healthy

Keywords

Structure Learning; Cognitive Flexibility; Neural Activation; Intervention; Predictive statistics

Outcome Measures

Primary Outcomes (38)

• Structure Learning Outcome 1 - Performance Index (PI) Relative

  Performance index (PI) is the minimum overlap between the distribution of participant responses and the distribution of presented targets per context. The overall PI is then the average of the performance indices across contexts. The PI relative is a normalised PI measure that quantifies participant's performance relative to random guessing.

  Across two weeks of intervention

• Structure Learning Outcome 2 - Strategy choice

  Strategy choice is the difference between the Kullback-Leibler (KL) divergence from model matching to the response-based model and the KL divergence from model maximization to the response-based model. Negative strategy choice values indicate a strategy closer to matching, whereas positive values indicate a strategy closer to maximization.

  Across two weeks of intervention

• Structure Learning Outcome 3 - Strategy index (measured by the integral curve difference)

  Derived by calculating the integral of each participant's strategy curve and subtracting it from the integral of the exact matching curve, as defined by model matching across training.

  Across two weeks of intervention.

• Structure Learning Outcome 4 - Learning rate

  Rate of learning across the training sessions or days

  Across two weeks of intervention.

• Structure Learning Outcome 5 - Strategy shifting rate

  Rate of strategy change across the training sessions or days

  Across two weeks of intervention.

• Cognitive Flexibility Outcome 1 - Wisconsin Card Sorting Test (WCST): Proportion of perseverative errors

  Proportion of error trials where participants incorrectly choose a deck based on the rule from the previous set

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 2 - Wisconsin Card Sorting Test (WCST): Learning rate

  This is a parameter extracted from a sequential learning model fitted to WCST data and represents how quickly participants updated their beliefs about the values associated with choices following respective negative and positive feedback.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 3 - Wisconsin Card Sorting Test (WCST): Decision consistency

  This is a parameter extracted from a sequential learning model fitted to WCST data and influenced the estimated probability of choosing a specific stimulus per trial. Larger values indicated increased exploitation whereas lower values indicated increased exploration.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 4 - Intra/Extra-Dimensional Set Shifting (IED): Extra-dimensional set errors

  The number of times that the subject failed to select the stimulus compatible with the current rule on the stage where the extra-dimensional shift occurs.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 5 - Intra/Extra-Dimensional Set Shifting (IED): Extra-dimensional reversal errors

  This is the total number of errors made in Stage 9. The lower the better.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 6 - Intra/Extra-Dimensional Set Shifting (IED): Learning rate

  Parameter extracted from Feature Reinforcement Model

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 7 - Intra/Extra-Dimensional Set Shifting (IED): Choice determinism

  Parameter extracted from Feature Reinforcement Model

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 8 - Intra/Extra-Dimensional Set Shifting (IED): Dimension primacy

  Parameter extracted from Feature Reinforcement Model

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 9 - Probabilistic Reversal Learning (PRL): Perseveration

  Number of trials until the participant updates their response after the rule reversal.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 10 - Probabilistic Reversal Learning (PRL): Switch probability

  Number of switches in the participant responses following negative feedback (i.e. trap trials: when the less likely object is correct).

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 11 - Probabilistic Reversal Learning (PRL): Trials to criterion in block 2 (Reversal)

  Number of trials before participant reach defined learning criterion during the Reversal phase

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 12 - Probabilistic Reversal Learning (PRL): Learning rates

  Parameter extracted from the Reinforcement Learning Model based on the Rescorla-Wagner rule. Two learning rates i.e., one for positive error/reward and one for negative error/punishment can be extracted from the model.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 13 - Probabilistic Reversal Learning (PRL): Reinforcement sensitivity

  Parameter extracted from the Reinforcement Learning Model based on the Rescorla-Wagner rule.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 14 - Probabilistic Reversal Learning (PRL): Stickiness

  Parameter extracted from the Reinforcement Learning Model based on the Rescorla-Wagner rule.

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 15 - Task Set Switching - Where (TSS-Where): Task switch cost in RT

  Switch cost in terms of reaction time (ms) is defined as the difference between the mean reaction time across switch trials with that of repeat trials

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 16 - Task Set Switching - Where (TSS-Where): Task switch cost in Accuracy

  Switch cost in terms of accuracy is defined as the difference between the accuracy in switch trials with that of repeat trials

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 17 - Task Set Switching - What (TSS-What): Task switch cost in RT

  Switch cost in terms of reaction time (ms) is defined as the difference between the mean reaction time across switch trials with that of repeat trials

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 18 - Task Set Switching - What (TSS-What): Task switch cost in Accuracy

  Switch cost in terms of accuracy is defined as the difference between the accuracy in switch trials with that of repeat trials

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 19 - Colour Shape Task (CST): Task switch cost in RT

  Switch cost in terms of reaction time (ms) is defined as the difference between the mean reaction time across switch trials with that of repeat trials. Used for matching between Control and Training groups

  Baseline (pre-intervention)

• Cognitive Flexibility Outcome 20 - Colour Shape Task (CST): Task switch cost in Accuracy

  Switch cost in terms of accuracy is defined as the difference between the accuracy in switch trials with that of repeat trials. Used for matching between Control and Training groups

  Baseline (pre-intervention)

• Cognitive Flexibility Outcome 21 - Colour Shape Task (CST): Task switch cost in RT

  Switch cost in terms of reaction time (ms) is defined as the difference between the mean reaction time across switch trials with that of repeat trials

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 22 - Colour Shape Task (CST): Task switch cost in Accuracy

  Switch cost in terms of reaction time (ms) is defined as the difference between the mean reaction time across switch trials with that of repeat trials

  One week after post-MRI scan (post-intervention)

• Cognitive Flexibility Outcome 23 - Colour Shape Task (CST): Change in task switch cost in RT

  Change in switch cost as indexed by reaction time between pre and post intervention

  Baseline (pre-intervention) and 8 weeks (after post-MRI scan, post-intervention)

• Cognitive Flexibility Outcome 24 - Colour Shape Task (CST): Change in task switch cost in accuracy

  Change in switch cost as indexed by accuracy between pre and post intervention

  Baseline (pre-intervention) and 8 weeks (after post-MRI scan, post-intervention)

• Cognitive Flexibility Outcome 25 - Trail Making Test (TMT): Ratio between time to complete Trail B to Trail A

  Participants are asked to complete Trail A (draw lines to connect numbered items in sequence) and subsequently Trail B (draw lines to connect a set of items alternating between letters and numbers) as quickly and as accurately as possible. We then take the ratio of the completion time of Trail B to Trail A to examine completion time of Trail B relative to Trail A.

  One week after post-MRI scan (post-intervention)

• Working Memory Outcome 1 - Backwards Digit Span (BDS): Total items correct

  Number of digit sequences that participant successfully recalled in the reverse order

  One week after post-MRI scan (post-intervention)

• Working Memory Outcome 2 - Reading Span (RS): Number of correctly recalled digits

  Total number of correctly recalled digits across the different trials

  One week after post-MRI scan (post-intervention)

• Working Memory Outcome 3 - Spatial Working Memory (SWM): Between search errors

  The number of times the subject incorrectly revisits a box in which a token has previously been found. Calculated across all assessed four, six and eight token trials.

  One week after post-MRI scan (post-intervention)

• Working Memory Outcome 4 - Spatial Working Memory (SWM): Strategy score

  The number of times a subject begins a new search pattern from the same box they started with previously. If they always begin a search from the same starting point we infer that the subject is employing a planned strategy for finding the tokens. Therefore a low score indicates high strategy use (1 = they always begin the search from the same box), a high score indicates that they are beginning their searches from many different boxes. Calculated across assessed trials with 6 tokens or more.

  One week after post-MRI scan (post-intervention)

• Inhibition Outcome 1 - Stroop Task: Interference measure for RT

  Difference between response latency of all correct congruent trials and response latency of all correct incongruent trials

  One week after post-MRI scan (post-intervention)

• Inhibition Outcome 2 - Stroop Task: Interference measure for accuracy

  Difference between proportion correct of all congruent trials and proportion correct of all incongruent trials

  One week after post-MRI scan (post-intervention)

• Inhibition Outcome 3 - Stop Signal Task (SST): Probability of Go responses on Stop trials

  Probability of reacting in Stop Signal Trials

  One week after post-MRI scan (post-intervention)

• Inhibition Outcome 4 - Stop Signal Task (SST): Stop Signal Reaction Time

  Estimation of the covert stop signal reaction time (SSRT) (the time required to stop the initiated go-process) using the integration method (see Verbruggen et al, 2019). The slower the SSRT the more difficult to stop the go-process. The faster the SSRT the easier to stop the go-process.

  One week after post-MRI scan (post-intervention)

Secondary Outcomes (48)

• Perceived Stress Outcome 1 - Perceived Stress Scale (PSS): Perceived Stress Scale Score

  One week after post-MRI scan (post-intervention)

• Sleep Quality Outcome 1 - Pittsburgh Sleep Quality Index (PSQI): Global PSQI Score

  One week after post-MRI scan (post-intervention)

• Empathy Outcome 1 - Empathy Quotient (EQ): Empathy Quotient Score

  Baseline (pre-intervention) and 8 weeks (after post-MRI scan, post-intervention)

• Language and Numeracy Outcome 1 - Woodcock Johnson IV (WJIV): Scores from Test 1 Letter-Word Identification

  One week before pre-MRI scan (pre-intervention)

• Language and Numeracy Outcome 2 - Woodcock Johnson IV (WJIV): Scores from Test 2 Applied Problems

  One week before pre-MRI scan (pre-intervention)

Other Outcomes (10)

• Edinburgh Handedness Inventory Score: Total handedness score

  Baseline (pre-intervention)

• Mood and State Outcome 1 - Self-Assessment Manikin (SAM)

  Daily across two weeks of intervention.

• Mood and State Outcome 2 - Self-Assessment Manikin (SAM)

  Four weeks from baseline, at pre-mri scan (pre-intervention)

Study Arms (2)

Structure Learning Training

EXPERIMENTAL

The intervention group will receive structure learning training that tap on their ability to extract patterns from prior stimuli presentations to make predictions. Participants will be presented with visual sequences of symbols determined by frequency statistics and upon mastery, more complex context-based statistics. No feedback will be provided and participants are trained in an adaptive manner

Behavioral: Structure Learning Training

Passive Control

NO INTERVENTION

The passive control group will not receive any intervention but will receive the same pre-post cognitive-behavioural and neuroimaging intervention assessments.

Interventions

Structure Learning Training BEHAVIORAL

Participants will undergo 12 sessions of structure learning training lasting about 1 hour each along with 2 testing sessions interspersed within the 12 training sessions. Each session will be conducted in a remote-guided manner with an approximate 1-day gap in between sessions. The entire training will last around two weeks

Structure Learning Training

Eligibility Criteria

• Age: 18 Years - 55 Years
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64)

You may qualify if:

• Healthy volunteer (male of female) between 18 and 55 years who gave written informed consent to participate

You may not qualify if:

• Current and/or prior history of learning disabilities
• Current and/or prior history of neurological disorder
• Current and/or prior history of psychiatric disorder
• Current and/or prior history of cardiovascular disorder
• Predominantly left-handed
• Contraindications for MRI (e.g., pacemakers, implanted pumps, metal objects in the body)
• Claustrophobic
• Pregnancy (females)
• Lactation (females)
• Pronounced visual or auditory impairments

Sponsors & Collaborators

• Nanyang Technological University: lead
• University of Cambridge: collaborator

Study Sites (1)

Centre for Lifelong Learning and Individualised Cognition (CLIC), Nanyang Technological University

Singapore, 637335, Singapore

Location

Related Publications (17)

• Friedman NP, Miyake A. Unity and diversity of executive functions: Individual differences as a window on cognitive structure. Cortex. 2017 Jan;86:186-204. doi: 10.1016/j.cortex.2016.04.023. Epub 2016 May 10.

  PMID: 27251123 BACKGROUND

• Dajani DR, Uddin LQ. Demystifying cognitive flexibility: Implications for clinical and developmental neuroscience. Trends Neurosci. 2015 Sep;38(9):571-8. doi: 10.1016/j.tins.2015.07.003.

  PMID: 26343956 BACKGROUND

• Uddin LQ. Cognitive and behavioural flexibility: neural mechanisms and clinical considerations. Nat Rev Neurosci. 2021 Mar;22(3):167-179. doi: 10.1038/s41583-021-00428-w. Epub 2021 Feb 3.

  PMID: 33536614 BACKGROUND

• Karbach J, Kray J. How useful is executive control training? Age differences in near and far transfer of task-switching training. Dev Sci. 2009 Nov;12(6):978-90. doi: 10.1111/j.1467-7687.2009.00846.x.

  PMID: 19840052 BACKGROUND

• Buttelmann F, Karbach J. Development and Plasticity of Cognitive Flexibility in Early and Middle Childhood. Front Psychol. 2017 Jun 20;8:1040. doi: 10.3389/fpsyg.2017.01040. eCollection 2017.

  PMID: 28676784 BACKGROUND

• Kloo D, Perner J. Training transfer between card sorting and false belief understanding: helping children apply conflicting descriptions. Child Dev. 2003 Nov-Dec;74(6):1823-39. doi: 10.1046/j.1467-8624.2003.00640.x.

  PMID: 14669898 BACKGROUND

• Wang R, Shen Y, Tino P, Welchman AE, Kourtzi Z. Learning predictive statistics from temporal sequences: Dynamics and strategies. J Vis. 2017 Oct 1;17(12):1. doi: 10.1167/17.12.1.

  PMID: 28973111 BACKGROUND

• Wang R, Shen Y, Tino P, Welchman AE, Kourtzi Z. Learning Predictive Statistics: Strategies and Brain Mechanisms. J Neurosci. 2017 Aug 30;37(35):8412-8427. doi: 10.1523/JNEUROSCI.0144-17.2017. Epub 2017 Jul 31.

  PMID: 28760866 BACKGROUND

• Karlaftis VM, Wang R, Shen Y, Tino P, Williams G, Welchman AE, Kourtzi Z. White-Matter Pathways for Statistical Learning of Temporal Structures. eNeuro. 2018 Jul 17;5(3):ENEURO.0382-17.2018. doi: 10.1523/ENEURO.0382-17.2018. eCollection 2018 May-Jun.

  PMID: 30027110 BACKGROUND

• Karlaftis VM, Giorgio J, Vertes PE, Wang R, Shen Y, Tino P, Welchman AE, Kourtzi Z. Multimodal imaging of brain connectivity reveals predictors of individual decision strategy in statistical learning. Nat Hum Behav. 2019 Mar 1;3:297-307. doi: 10.1038/s41562-018-0503-4.

  PMID: 30873437 BACKGROUND

• Giorgio J, Karlaftis VM, Wang R, Shen Y, Tino P, Welchman A, Kourtzi Z. Functional brain networks for learning predictive statistics. Cortex. 2018 Oct;107:204-219. doi: 10.1016/j.cortex.2017.08.014. Epub 2017 Aug 18.

  PMID: 28923313 BACKGROUND

• Green CS, Bavelier D. Learning, attentional control, and action video games. Curr Biol. 2012 Mar 20;22(6):R197-206. doi: 10.1016/j.cub.2012.02.012.

  PMID: 22440805 BACKGROUND

• Bock AM, Cartwright KB, McKnight PE, Patterson AB, Shriver AG, Leaf BM, Mohtasham MK, Vennergrund KC, Pasnak R. Patterning, Reading, and Executive Functions. Front Psychol. 2018 Sep 25;9:1802. doi: 10.3389/fpsyg.2018.01802. eCollection 2018.

  PMID: 30319500 BACKGROUND

• Yeniad N, Malda M, Mesman J, Van IJzendoorn MH, Pieper S. Shifting ability predicts math and reading performance in children: A meta-analytical study. Learning and Individual Differences. 2013; 23:1-9.

  BACKGROUND

• Bock AM, Gallaway KC, Hund AM. Specifying links between executive functioning and theory of mind during middle childhood: Cognitive flexibility predicts social understanding. Journal of Cognition and Development. 2015;16(3):509-21.

  BACKGROUND

• Papic MM, Mulligan JT, Mitchelmore MC. Assessing the development of preschoolers' mathematical patterning. Journal for Research in Mathematics Education. 2011;42(3):237-68.

  BACKGROUND

• Liu CL, Cheng X, Choo BL, Hong M, Teo JL, Koo WL, Tan JYJ, Ubrani MB, Suckling J, Gulyas B, Leong V, Kourtzi Z, Sahakian B, Robbins T, Chen AS. Potential cognitive and neural benefits of a computerised cognitive training programme based on Structure Learning in healthy adults: study protocol for a randomised controlled trial. Trials. 2023 Aug 11;24(1):517. doi: 10.1186/s13063-023-07551-2.

  PMID: 37568212 DERIVED

Study Officials

• Annabel Chen, PhD

  Nanyang Technological University

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Purpose: BASIC SCIENCE
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Professor

Model Details: Participants are pseudo-randomly assigned (matched for age, sex, intelligence and flexibility) to one of the following intervention groups: (1) Structure Learning and (2) No training (Passive Control)

Study Record Dates

• First Submitted: October 7, 2022
• First Posted: November 10, 2022
• Study Start: June 3, 2022
• Primary Completion: June 4, 2023
• Study Completion: June 4, 2023
• Last Updated: May 28, 2025

Record last verified: 2024-02

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, SAP, ANALYTIC CODE
• Time Frame: As soon as data collection commenced
• Access Criteria: Researchers may apply for access to data

Locations

SG (1)